#826446 (C++17(gcc12)) No.5011 Better Mo's Algorithm is Needed!! (Weighted)

• 問題: No.5011 Better Mo's Algorithm is Needed!! (Weighted)
• ユーザー: NyaanNyaan
• 提出日時: 2022-12-17 13:46:35
• 言語: C++17(gcc12) (gcc 12.3.0 + boost 1.87.0)
• 結果: AC
• 実行時間: 4,909 ms / 5,000 ms
• コード長: 18,899 bytes
• コンパイル時間: 5,142 ms
• 実行使用メモリ: 10,388 KB
• スコア: 43,401,326,439
• 最終ジャッジ日時: 2022-12-17 14:05:40
• 合計ジャッジ時間: 644,735 ms
• ジャッジサーバーID （参考情報）: judge16 / judge11

ファイルパターン	結果
other	AC * 120

ソースコード

/**
 *  date : 2022-12-17 13:45:37
 */

#define NDEBUG
using namespace std;

// intrinstic
#include <immintrin.h>

#include <algorithm>
#include <array>
#include <bitset>
#include <cassert>
#include <cctype>
#include <cfenv>
#include <cfloat>
#include <chrono>
#include <cinttypes>
#include <climits>
#include <cmath>
#include <complex>
#include <cstdarg>
#include <cstddef>
#include <cstdint>
#include <cstdio>
#include <cstdlib>
#include <cstring>
#include <deque>
#include <fstream>
#include <functional>
#include <initializer_list>
#include <iomanip>
#include <ios>
#include <iostream>
#include <istream>
#include <iterator>
#include <limits>
#include <list>
#include <map>
#include <memory>
#include <new>
#include <numeric>
#include <ostream>
#include <queue>
#include <random>
#include <set>
#include <sstream>
#include <stack>
#include <streambuf>
#include <string>
#include <tuple>
#include <type_traits>
#include <typeinfo>
#include <unordered_map>
#include <unordered_set>
#include <utility>
#include <vector>

// utility
namespace Nyaan {
using ll = long long;
using i64 = long long;
using u64 = unsigned long long;
using i128 = __int128_t;
using u128 = __uint128_t;

template <typename T>
using V = vector<T>;
template <typename T>
using VV = vector<vector<T>>;
using vi = vector<int>;
using vl = vector<long long>;
using vd = V<double>;
using vs = V<string>;
using vvi = vector<vector<int>>;
using vvl = vector<vector<long long>>;

template <typename T, typename U>
struct P : pair<T, U> {
  template <typename... Args>
  P(Args... args) : pair<T, U>(args...) {}

  using pair<T, U>::first;
  using pair<T, U>::second;

  P &operator+=(const P &r) {
    first += r.first;
    second += r.second;
    return *this;
  }
  P &operator-=(const P &r) {
    first -= r.first;
    second -= r.second;
    return *this;
  }
  P &operator*=(const P &r) {
    first *= r.first;
    second *= r.second;
    return *this;
  }
  template <typename S>
  P &operator*=(const S &r) {
    first *= r, second *= r;
    return *this;
  }
  P operator+(const P &r) const { return P(*this) += r; }
  P operator-(const P &r) const { return P(*this) -= r; }
  P operator*(const P &r) const { return P(*this) *= r; }
  template <typename S>
  P operator*(const S &r) const {
    return P(*this) *= r;
  }
  P operator-() const { return P{-first, -second}; }
};

using pl = P<ll, ll>;
using pi = P<int, int>;
using vp = V<pl>;

constexpr int inf = 1001001001;
constexpr long long infLL = 4004004004004004004LL;

template <typename T>
int sz(const T &t) {
  return t.size();
}

template <typename T, typename U>
inline bool amin(T &x, U y) {
  return (y < x) ? (x = y, true) : false;
}
template <typename T, typename U>
inline bool amax(T &x, U y) {
  return (x < y) ? (x = y, true) : false;
}

template <typename T>
inline T Max(const vector<T> &v) {
  return *max_element(begin(v), end(v));
}
template <typename T>
inline T Min(const vector<T> &v) {
  return *min_element(begin(v), end(v));
}
template <typename T>
inline long long Sum(const vector<T> &v) {
  return accumulate(begin(v), end(v), 0LL);
}

template <typename T>
int lb(const vector<T> &v, const T &a) {
  return lower_bound(begin(v), end(v), a) - begin(v);
}
template <typename T>
int ub(const vector<T> &v, const T &a) {
  return upper_bound(begin(v), end(v), a) - begin(v);
}

constexpr long long TEN(int n) {
  long long ret = 1, x = 10;
  for (; n; x *= x, n >>= 1) ret *= (n & 1 ? x : 1);
  return ret;
}

template <typename T, typename U>
pair<T, U> mkp(const T &t, const U &u) {
  return make_pair(t, u);
}

template <typename T>
vector<T> mkrui(const vector<T> &v, bool rev = false) {
  vector<T> ret(v.size() + 1);
  if (rev) {
    for (int i = int(v.size()) - 1; i >= 0; i--) ret[i] = v[i] + ret[i + 1];
  } else {
    for (int i = 0; i < int(v.size()); i++) ret[i + 1] = ret[i] + v[i];
  }
  return ret;
};

template <typename T>
vector<T> mkuni(const vector<T> &v) {
  vector<T> ret(v);
  sort(ret.begin(), ret.end());
  ret.erase(unique(ret.begin(), ret.end()), ret.end());
  return ret;
}

template <typename F>
vector<int> mkord(int N,F f) {
  vector<int> ord(N);
  iota(begin(ord), end(ord), 0);
  sort(begin(ord), end(ord), f);
  return ord;
}

template <typename T>
vector<int> mkinv(vector<T> &v) {
  int max_val = *max_element(begin(v), end(v));
  vector<int> inv(max_val + 1, -1);
  for (int i = 0; i < (int)v.size(); i++) inv[v[i]] = i;
  return inv;
}

vector<int> mkiota(int n) {
  vector<int> ret(n);
  iota(begin(ret), end(ret), 0);
  return ret;
}

template <typename T>
T mkrev(const T &v) {
  T w{v};
  reverse(begin(w), end(w));
  return w;
}

template <typename T>
bool nxp(vector<T> &v) {
  return next_permutation(begin(v), end(v));
}

template <typename T>
using minpq = priority_queue<T, vector<T>, greater<T>>;

}  // namespace Nyaan

// bit operation
namespace Nyaan {
__attribute__((target("popcnt"))) inline int popcnt(const u64 &a) {
  return _mm_popcnt_u64(a);
}
inline int lsb(const u64 &a) { return a ? __builtin_ctzll(a) : 64; }
inline int ctz(const u64 &a) { return a ? __builtin_ctzll(a) : 64; }
inline int msb(const u64 &a) { return a ? 63 - __builtin_clzll(a) : -1; }
template <typename T>
inline int gbit(const T &a, int i) {
  return (a >> i) & 1;
}
template <typename T>
inline void sbit(T &a, int i, bool b) {
  if (gbit(a, i) != b) a ^= T(1) << i;
}
constexpr long long PW(int n) { return 1LL << n; }
constexpr long long MSK(int n) { return (1LL << n) - 1; }
}  // namespace Nyaan

// inout
namespace Nyaan {

template <typename T, typename U>
ostream &operator<<(ostream &os, const pair<T, U> &p) {
  os << p.first << " " << p.second;
  return os;
}
template <typename T, typename U>
istream &operator>>(istream &is, pair<T, U> &p) {
  is >> p.first >> p.second;
  return is;
}

template <typename T>
ostream &operator<<(ostream &os, const vector<T> &v) {
  int s = (int)v.size();
  for (int i = 0; i < s; i++) os << (i ? " " : "") << v[i];
  return os;
}
template <typename T>
istream &operator>>(istream &is, vector<T> &v) {
  for (auto &x : v) is >> x;
  return is;
}

istream &operator>>(istream &is, __int128_t &x) {
  string S;
  is >> S;
  x = 0;
  int flag = 0;
  for (auto &c : S) {
    if (c == '-') {
      flag = true;
      continue;
    }
    x *= 10;
    x += c - '0';
  }
  if (flag) x = -x;
  return is;
}

istream &operator>>(istream &is, __uint128_t &x) {
  string S;
  is >> S;
  x = 0;
  for (auto &c : S) {
    x *= 10;
    x += c - '0';
  }
  return is;
}

ostream &operator<<(ostream &os, __int128_t x) {
  if (x == 0) return os << 0;
  if (x < 0) os << '-', x = -x;
  string S;
  while (x) S.push_back('0' + x % 10), x /= 10;
  reverse(begin(S), end(S));
  return os << S;
}
ostream &operator<<(ostream &os, __uint128_t x) {
  if (x == 0) return os << 0;
  string S;
  while (x) S.push_back('0' + x % 10), x /= 10;
  reverse(begin(S), end(S));
  return os << S;
}

void in() {}
template <typename T, class... U>
void in(T &t, U &...u) {
  cin >> t;
  in(u...);
}

void out() { cout << "\n"; }
template <typename T, class... U, char sep = ' '>
void out(const T &t, const U &...u) {
  cout << t;
  if (sizeof...(u)) cout << sep;
  out(u...);
}

void outr() {}
template <typename T, class... U, char sep = ' '>
void outr(const T &t, const U &...u) {
  cout << t;
  outr(u...);
}

struct IoSetupNya {
  IoSetupNya() {
    cin.tie(nullptr);
    ios::sync_with_stdio(false);
    cout << fixed << setprecision(15);
    cerr << fixed << setprecision(7);
  }
} iosetupnya;

}  // namespace Nyaan

// debug

#ifdef NyaanDebug
#define trc(...) (void(0))
#else
#define trc(...) (void(0))
#endif

#ifdef NyaanLocal
#define trc2(...) (void(0))
#else
#define trc2(...) (void(0))
#endif

// macro
#define each(x, v) for (auto&& x : v)
#define each2(x, y, v) for (auto&& [x, y] : v)
#define all(v) (v).begin(), (v).end()
#define rep(i, N) for (long long i = 0; i < (long long)(N); i++)
#define repr(i, N) for (long long i = (long long)(N)-1; i >= 0; i--)
#define rep1(i, N) for (long long i = 1; i <= (long long)(N); i++)
#define repr1(i, N) for (long long i = (N); (long long)(i) > 0; i--)
#define reg(i, a, b) for (long long i = (a); i < (b); i++)
#define regr(i, a, b) for (long long i = (b)-1; i >= (a); i--)
#define fi first
#define se second
#define ini(...)   \
  int __VA_ARGS__; \
  in(__VA_ARGS__)
#define inl(...)         \
  long long __VA_ARGS__; \
  in(__VA_ARGS__)
#define ins(...)      \
  string __VA_ARGS__; \
  in(__VA_ARGS__)
#define in2(s, t)                           \
  for (int i = 0; i < (int)s.size(); i++) { \
    in(s[i], t[i]);                         \
  }
#define in3(s, t, u)                        \
  for (int i = 0; i < (int)s.size(); i++) { \
    in(s[i], t[i], u[i]);                   \
  }
#define in4(s, t, u, v)                     \
  for (int i = 0; i < (int)s.size(); i++) { \
    in(s[i], t[i], u[i], v[i]);             \
  }
#define die(...)             \
  do {                       \
    Nyaan::out(__VA_ARGS__); \
    return;                  \
  } while (0)

namespace Nyaan {
void solve();
}
int main() { Nyaan::solve(); }

//

struct log_table {
  double l[65536];
  constexpr log_table() : l() {
    unsigned long long x = 88172645463325252ULL;
    double log_u64max = log(2) * 64;
    for (int i = 0; i < 65536; i++) {
      x = x ^ (x << 7);
      x = x ^ (x >> 9);
      l[i] = log(double(x)) - log_u64max;
    }
  }
};



using namespace std;

namespace fastio {
static constexpr int SZ = 1 << 17;
char inbuf[SZ], outbuf[SZ];
int in_left = 0, in_right = 0, out_right = 0;

struct Pre {
  char num[40000];
  constexpr Pre() : num() {
    for (int i = 0; i < 10000; i++) {
      int n = i;
      for (int j = 3; j >= 0; j--) {
        num[i * 4 + j] = n % 10 + '0';
        n /= 10;
      }
    }
  }
} constexpr pre;

inline void load() {
  int len = in_right - in_left;
  memmove(inbuf, inbuf + in_left, len);
  in_right = len + fread(inbuf + len, 1, SZ - len, stdin);
  in_left = 0;
}

inline void flush() {
  fwrite(outbuf, 1, out_right, stdout);
  out_right = 0;
}

inline void skip_space() {
  if (in_left + 32 > in_right) load();
  while (inbuf[in_left] <= ' ') in_left++;
}

inline void rd(char& c) {
  if (in_left + 32 > in_right) load();
  c = inbuf[in_left++];
}
template <typename T>
inline void rd(T& x) {
  if (in_left + 32 > in_right) load();
  char c;
  do c = inbuf[in_left++];
  while (c < '-');
  [[maybe_unused]] bool minus = false;
  if constexpr (is_signed<T>::value == true) {
    if (c == '-') minus = true, c = inbuf[in_left++];
  }
  x = 0;
  while (c >= '0') {
    x = x * 10 + (c & 15);
    c = inbuf[in_left++];
  }
  if constexpr (is_signed<T>::value == true) {
    if (minus) x = -x;
  }
}
inline void rd() {}
template <typename Head, typename... Tail>
inline void rd(Head& head, Tail&... tail) {
  rd(head);
  rd(tail...);
}

inline void wt(char c) {
  if (out_right > SZ - 32) flush();
  outbuf[out_right++] = c;
}
inline void wt(bool b) {
  if (out_right > SZ - 32) flush();
  outbuf[out_right++] = b ? '1' : '0';
}
inline void wt(const string &s) {
  if (out_right + s.size() > SZ - 32) flush();
  memcpy(outbuf + out_right, s.data(), sizeof(char) * s.size());
  out_right += s.size();
}
template <typename T>
inline void wt(T x) {
  if (out_right > SZ - 32) flush();
  if (!x) {
    outbuf[out_right++] = '0';
    return;
  }
  if constexpr (is_signed<T>::value == true) {
    if (x < 0) outbuf[out_right++] = '-', x = -x;
  }
  int i = 12;
  char buf[16];
  while (x >= 10000) {
    memcpy(buf + i, pre.num + (x % 10000) * 4, 4);
    x /= 10000;
    i -= 4;
  }
  if (x < 100) {
    if (x < 10) {
      outbuf[out_right] = '0' + x;
      ++out_right;
    } else {
      uint32_t q = (uint32_t(x) * 205) >> 11;
      uint32_t r = uint32_t(x) - q * 10;
      outbuf[out_right] = '0' + q;
      outbuf[out_right + 1] = '0' + r;
      out_right += 2;
    }
  } else {
    if (x < 1000) {
      memcpy(outbuf + out_right, pre.num + (x << 2) + 1, 3);
      out_right += 3;
    } else {
      memcpy(outbuf + out_right, pre.num + (x << 2), 4);
      out_right += 4;
    }
  }
  memcpy(outbuf + out_right, buf + i + 4, 12 - i);
  out_right += 12 - i;
}
inline void wt() {}
template <typename Head, typename... Tail>
inline void wt(Head&& head, Tail&&... tail) {
  wt(head);
  wt(forward<Tail>(tail)...);
}
template <typename... Args>
inline void wtn(Args&&... x) {
  wt(forward<Args>(x)...);
  wt('\n');
}

struct Dummy {
  Dummy() { atexit(flush); }
} dummy;

}  // namespace fastio
using fastio::rd;
using fastio::skip_space;
using fastio::wt;
using fastio::wtn;


namespace my_rand {
using i64 = long long;
using u64 = unsigned long long;

// [0, 2^64 - 1)
u64 rng() {
  static u64 _x =
      u64(chrono::duration_cast<chrono::nanoseconds>(
              chrono::high_resolution_clock::now().time_since_epoch())
              .count()) *
      10150724397891781847ULL;
  _x ^= _x << 7;
  return _x ^= _x >> 9;
}

// [l, r]
i64 rng(i64 l, i64 r) {
  assert(l <= r);
  return l + rng() % (r - l + 1);
}

// [l, r)
i64 randint(i64 l, i64 r) {
  assert(l < r);
  return l + rng() % (r - l);
}

// choose n numbers from [l, r) without overlapping
vector<i64> randset(i64 l, i64 r, i64 n) {
  assert(l <= r && n <= r - l);
  unordered_set<i64> s;
  for (i64 i = n; i; --i) {
    i64 m = randint(l, r + 1 - i);
    if (s.find(m) != s.end()) m = r - i;
    s.insert(m);
  }
  vector<i64> ret;
  for (auto& x : s) ret.push_back(x);
  return ret;
}

// [0.0, 1.0)
double rnd() { return rng() * 5.42101086242752217004e-20; }

template <typename T>
void randshf(vector<T>& v) {
  int n = v.size();
  for (int i = 1; i < n; i++) swap(v[i], v[randint(0, i + 1)]);
}

}  // namespace my_rand

using my_rand::randint;
using my_rand::randset;
using my_rand::randshf;
using my_rand::rnd;
using my_rand::rng;



struct Timer {
  chrono::high_resolution_clock::time_point st;

  Timer() { reset(); }

  void reset() { st = chrono::high_resolution_clock::now(); }

  chrono::milliseconds::rep elapsed() {
    auto ed = chrono::high_resolution_clock::now();
    return chrono::duration_cast<chrono::milliseconds>(ed - st).count();
  }
};
using namespace Nyaan;

// ref : https://codeforces.com/blog/entry/61203
const int segDelta[2][2] = {{2, 1}, {3, 0}};
const int rotateDelta[4] = {3, 0, 0, 1};
int64_t hilbertOrder(int x, int y, int pow, int rotate) {
  if (pow == 0) return 0;
  int hpow = 1 << (pow - 1);
  bool xflag = x < hpow, yflag = y < hpow;
  int nx = xflag ? x : x - hpow;
  int ny = yflag ? y : y - hpow;
  int seg = (rotate + segDelta[xflag][yflag]) & 3;
  int nrot = (rotate + rotateDelta[seg]) & 3;
  int64_t subSquareSize = int64_t(1) << (2 * pow - 2);
  int64_t ans = int64_t(seg) << (2 * pow - 2);
  int64_t add = hilbertOrder(nx, ny, pow - 1, nrot);
  ans += (seg == 1 || seg == 2) ? add : (subSquareSize - add - 1);
  return ans;
}

constexpr int N = 200000;
constexpr int Q = 200000;
int rui[200003], p[200003], d[200003];
pair<int, int> xy[200003];
ll hil[200003];

void q() {
  Timer timer;

  /**/
  {
    int d1, d2, d3, d4;
    rd(d1, d2, d3, d4);
  }
  rep(i, N) rd(rui[i + 1]), rui[i + 1] += rui[i];
  rep1(i, Q) {
    rd(xy[i].fi, xy[i].se);
    xy[i].fi = rui[xy[i].fi - 1];
    xy[i].se = rui[xy[i].se];
  }
  //*/

  /**
  rep(i, N) rui[i + 1] = rng(1, 100), rui[i + 1] += rui[i];
  rep1(i, Q) {
    xy[i].fi = rng(1, N);
    xy[i].se = rng(1, N);
    if (xy[i].fi > xy[i].se) swap(xy[i].fi, xy[i].se);
    xy[i].fi = rui[xy[i].fi - 1];
    xy[i].se = rui[xy[i].se];
  }
  //*/

  constexpr int start = 0;

  trc2(timer.elapsed());
  ll pw = 1;
  while ((1LL << pw) <= rui[N]) pw++;
  rep(i, Q + 1) {
    p[i] = i;
    hil[i] = hilbertOrder(xy[i].fi, xy[i].se, pw, 0);
  }
  trc2(timer.elapsed());
  sort(p, p + Q + 1, [&](int i, int j) { return hil[i] < hil[j]; });

  /*
  [[maybe_unused]] int width =
      max<int>(1, 1.0 * rui[N] / max<double>(1.0, sqrt(Q * 2.0 / 3.0)));
  sort(p, p + Q + 1, [&](int a, int b) {
    auto [ax, ay] = xy[a];
    auto [bx, by] = xy[b];
    int ablock = ax / width, bblock = bx / width;
    if (ablock != bblock) return ablock < bblock;
    return ablock & 1 ? ay > by : ay < by;
  });
  */

  assert(p[0] == start);

  auto dist = [&](int i, int j) {
    auto& [x1, y1] = xy[i];
    auto& [x2, y2] = xy[j];
    return abs(x2 - x1) + abs(y2 - y1);
  };
  [[maybe_unused]] auto dsum = [&]() {
    ll ans = 0;
    rep(i, Q) ans += dist(p[i], p[i + 1]);
    return ans;
  };

  rep(i, Q) d[i] = dist(p[i], p[i + 1]);
  [[maybe_unused]] int rev_count = 0;

#ifdef NyaanLocal
  constexpr double end_time = 3600;
#else
  constexpr double end_time = 4900;
#endif

  rep(t, 2) {
    int dmax = 700;
    trc2(dmax, dsum(), timer.elapsed(), rev_count);
    for (int i = 1; i <= Q; i++) {
      if (i % 5 == 0 and timer.elapsed() > end_time) break;
      int pre1 = d[i - 1];
      int js = i + 1, je = min<int>(i + dmax, Q);
      for (int j = je - 1; j >= js; j--) {
        // (i-1) - (i), (j) - (j+1)
        int pre2 = d[j];
        int now1 = dist(p[i - 1], p[j]);
        int now2 = dist(p[i], p[j + 1]);
        if (now1 + now2 < pre1 + pre2) {
          reverse(p + i, p + j + 1);
          reverse(d + i, d + j);
          d[i - 1] = pre1 = now1;
          d[j] = now2;
#ifdef NyaanLocal
          rev_count++;
#endif
        }
      }
    }
  }

  rep(_, 1) {
    constexpr int dmax = 40;
    double start_time = timer.elapsed();
    if (start_time >= end_time) break;
    double inv_time = 1.0 / (end_time - start_time);
    double cur_time = start_time;
    double start_temp = 10000;
    double end_temp = rui[N] / N;
    double cur_temp = start_temp;
    double diff_temp = start_temp - end_temp;

    auto set_temp = [&]() {
      cur_temp = max(0.0, start_temp - cur_time * diff_temp * inv_time);
    };

    log_table rand_log;
    int iter = 0;
    for (;; iter++) {
      if ((iter & 0xFFFF) == 0) {
#ifdef NyaanLocal
        if ((iter & 0xFFFFFF) == 0)
          trc2(dsum(), timer.elapsed(), iter, rev_count);
#endif
        if ((cur_time = timer.elapsed()) > end_time) break;
        set_temp();
      }
      // [1, Q - 1]
      int i = rng() % (Q - 1) + 1;
      // [1, dmax - 1]
      int dj = rng() % (dmax - 1) + 1;
      int j = i + dj;
      if (Q <= j) continue;

      int pre1 = d[i - 1];
      int pre2 = d[j];
      int now1 = dist(p[i - 1], p[j]);
      int now2 = dist(p[i], p[j + 1]);
      int score = (pre1 + pre2) - (now1 + now2);

      if (score > cur_temp * rand_log.l[iter & 0xFFFF]) {
        reverse(p + i, p + j + 1);
        reverse(d + i, d + j);
        d[i - 1] = pre1 = now1;
        d[j] = now2;
#ifdef NyaanLocal
        rev_count++;
#endif
      }
    }
    trc2(dsum(), timer.elapsed(), iter, rev_count);
    cerr << "iter : " << iter << endl;
  }
  for (int i = 1; i <= Q; i++) {
    wt(p[i]);
    wt(" \n"[i == Q]);
  }
}

void Nyaan::solve() {
  int t = 1;
  // in(t);
  while (t--) q();
}